Using Blockchain for Improved Video Integrity Verification

Sarala Ghimire

doi:10.1109/TMM.2019.2925961

Return

Using Blockchain for Improved Video Integrity Verification

2020-01-01

PRE

Sarala Ghimire

Jae Young Choi

Bumshik Lee *

DOI：10.1109/TMM.2019.2925961

Original

Original request for help

Save

Abstract

En 中文

A video record plays a crucial role in providing evidence for crime scenes or road accidents. However, the main problem with the video record is that it is often vulnerable to various video tampering attacks. Although visual evidence is required to conduct an integrity verification before investigations, it is still difficult for human vision to detect a forgery. In this paper, we propose a novel video integrity verification method (IVM) that takes advantage of a blockchain framework. The proposed method employs an effective blockchain model in centralized video data, by combining a hash-based message authentication code and elliptic curve cryptography to verify the integrity of a video. In our method, video content with a predetermined size (segments) is key-hashed in a real-time manner and stored in a chronologically chained fashion, thus establishing an irrefutable database. The verification process applies the same procedure to the video segment and generates a hash value that can be compared with the hash in the blockchain. The proposed IVM is implemented on a PC environment, as well as on an accident data recorder-embedded system for verification. The experimental results show that the proposed method has better detection capabilities and robustness toward various kinds of tampering, such as copy-move, insert, and delete, as compared to other state-of-the-art methods. An analysis based on execution time along with an increase in the number of blocks within the blockchain shows a minimal overhead in the proposed method.

Keywords:

Blockchain

video integrity

elliptic curve cryptography (ECC)

HMAC

AI Summary

Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.